AI-enabled digital twins for thermal energy storage in renewable power systems: Multi-scale modelling, power-system integration, and co-simulation frameworks

Authors

• Manzoore Elahi M. Soudagar, Lishui University
• Anupama Routray, Siksha O Anusandhan (Deemed to be University)
• Yuvaraja Teekaraman, American International University
• V. Sandeep, JAIN (Deemed-to-be University)
• V. Mohanavel, Academy of Maritime Education and Training
• Alagar Karthick, Saveetha School of Engineering
• S. Ramesh, Universiti Malaya
• N. A.Abu Osman, Universiti Malaya
• Selvakumar Kathiresan, Sharda University

Document Type

Review

Publication Date

8-1-2026

Abstract

Achieving high levels of variable renewable energy (VRE) integration requires flexible resources that can operate across diverse temporal scales while ensuring system reliability, minimizing curtailment, and facilitating sector-coupled energy transfers. Thermal energy storage (TES) offers a cost-efficient, scalable solution for addressing these requirements; however, its value within the system increasingly hinges on sophisticated digitalization. This review consolidates recent advances in artificial intelligence (AI)-enabled digital twins and multi-scale modeling frameworks that enhance the monitoring, forecasting, and regulation of TES in renewable energy-dominated systems. We investigate the extent to which hybrid physics-machine learning (ML) models, reinforcement learning (RL) controllers, and physics-informed neural networks (PINNs) enhance TES efficacy in Concentrated Solar Power (CSP) facilities, district heating systems, industrial electrification, and power-to-heat (P2H) initiatives. Co-simulation methodologies that amalgamate TES models with unit commitment (UC), economic dispatch (ED), and capacity-expansion instruments are rigorously evaluated. Principal challenges, including data interoperability, cyber-physical security, model validation, and interpretability, are recognized as impediments to extensive implementation. The review culminates in a structured research agenda aligned with Sustainable Development Goals (SDGs) 7, 9, and 13, delineating priorities for advancing next-generation autonomous and sustainable TES systems.

Keywords

Digital twins, Multi-scale modelling, Power system co-simulation, Thermal energy storage, Unit commitment and economic dispatch

Publication Title

Electric Power Systems Research

ISSN

0378-7796

DOI

10.1016/j.epsr.2026.112947

Recommended Citation

Soudagar, Manzoore Elahi M.; Routray, Anupama; Teekaraman, Yuvaraja; Sandeep, V.; Mohanavel, V.; Karthick, Alagar; Ramesh, S.; Osman, N. A.Abu; and Kathiresan, Selvakumar, "AI-enabled digital twins for thermal energy storage in renewable power systems: Multi-scale modelling, power-system integration, and co-simulation frameworks" (2026). Research Publications (2026 to 2030). 85.
https://knova.um.edu.my/research_publications_2026_2030/85

Volume

257

Publisher

Elsevier